Optimization of the pretreatment with ultrasound
The experimental design for the ultrasound pretreatment is shown in Table 1. The independent variables were time and ultrasound amplitude (min and %, respectively), and the dependent variable was total monomeric anthocyanins content (mg of cyanidin-3-glucoside/L of juice). In Table 2, the results of the analysis of variance are shown. The coefficient of determination (R²) for the model was 0.96, and the F value (27.04) was higher than the tabulated value (4.39), which confirms that the mathematical model fits the experimental data. Regarding the model adjustment, it was found that the calculated F value (7.28) was lower than the tabulated value (9.28), which indicates that the lack of fit was insignificant (p > 0.05) and the mathematical model is valid.
Table 1
Experimental design matrix for the pretreatment with ultrasound before juicing by pressing.
Experiment | Amplitude (%) | Time (min) | TMA* (mg/L) |
1 | 25 | 3 | 388 |
2 | 25 | 11 | 373 |
3 | 75 | 3 | 440 |
4 | 75 | 11 | 484 |
5 | 14.6 | 7 | 365 |
6 | 85.4 | 7 | 511 |
7 | 50 | 1.3 | 437 |
8 | 50 | 12.7 | 459 |
9 | 50 | 7 | 513 |
10 | 50 | 7 | 499 |
11 | 50 | 7 | 495 |
12 | 50 | 7 | 504 |
Table 2
Analysis of variance (ANOVA) for blackberry juices' total monomeric anthocyanins content with the pretreatment with ultrasound.
Source | Sum of squares | Degrees of freedom | Medium square | F calc | F 0.05 |
Regression | 31739.65 | 5 | 6347.93 | 27.04 | 4.39 |
Residual | 1408.55 | 6 | 234.76 | | |
Lack of fit | 1238.54 | 3 | 412.85 | 7.28 | 9.28 |
Pure error | 170.01 | 3 | 56.67 | | |
Total | 33148.20 | 11 | | | |
R squared = 0.96 |
In Fig. 2, it was possible to determine that the conditions for pretreatment with ultrasound changed the content of anthocyanins in the juice. By analyzing the response surface (Fig. 2a), it can be seen that the best amplitude and time conditions were found around the central points. For the amplitude of ultrasound, values lower than 50% did not provide an increase in anthocyanins in the juice. However, when an amplitude above 50% is applied, it is possible to remove higher levels of anthocyanins. When the sonication lasts less than 5 min, this short pretreatment time hinders the extraction of anthocyanins. On the other hand, for more than 10 min of ultrasound, the content of anthocyanins decreases, probably due to the degradation of these compounds.
The Pareto chart (Fig. 2b) shows the effect of amplitude and time, confirming that both influenced the extraction of anthocyanins into the juice. In addition, the interaction between these parameters also significantly affected the process. Thus, the optimal conditions for pretreatment with ultrasound were those in the center of the response surface graph, employing 65% of the ultrasound amplitude for 7 minutes.
Effect of ultrasound pretreatment on anthocyanin content of the juice
When comparing the content of anthocyanins found in blackberry juices with and without ultrasound pretreatment (Fig. 3), it can be noted that ultrasound increased anthocyanins extraction for the juices of all cultivars, increasing their content by 31%, 31%, and 56% for the cultivars BRS Xingu, Guarani, and Xavante, respectively. Therefore, ultrasound can be used as a pretreatment before pressing to increase the removal of bioactive compounds from the fruit, making the juice more concentrated in anthocyanins.
It is well known that the presence of a liquid medium is required for ultrasound extraction of bioactive compounds. The sonication of the solvent leads to the formation of the cavitation effect, in which the generation and implosion of bubbles provide the extravasation of the cellular content of plant tissues (Chen et al. 2020). Since blackberry is composed of nearly 90% water (Moraes et al. 2020), it is possible to expect the cavitation effect without adding water, as reported here. Another fact that may have facilitated the extraction was the freezing that the sample was submitted before the pretreatment. Rapid freezing allows the formation of tiny, uniform ice crystals, causing minor damage to cell integrity. Slow freezing, which was used in this study, produces large ice crystals that cause significant damage to cell structure (Wu et al. 2017). Thus, the damage to the fruit's cellular structure generated by the slow freezing possibly contributed to the ultrasound extraction without the need to add water to the fruits.
It is important to mention that the implosion of cavitation bubbles generates microjets, which leads to surface flaking, erosion, and particle breakage, facilitating the release of anthocyanins from the intracellular environment to the in situ water in the fruit. Sonoporation is another effect reported in ultrasound-assisted extractions, which causes an increase in the permeability of cell membranes and the improvement in the release of cellular content (Chemat et al. 2017). Therefore, sonication can improve extraction efficiency, causing an increase in the mass transfer rate (Kumari et al. 2018).
Thus, the pretreatment of whole fruits without adding any solvent was feasible to increase the extraction from the fruits, resulting in juices with a higher concentration of anthocyanins.
Effect of ultrasound pretreatment on the composition of individual anthocyanins in blackberry juices
In Fig. 4 is shown the anthocyanin profile of blackberry juices without (control) and with ultrasound pretreatment. It can be seen that sonication had a positive effect on increasing the concentration of anthocyanins in the juices. In all cultivars, five anthocyanins were identified (Table 3), and the same compounds were identified in the ultrasound-treated juices and the controls. Four cyanidins with different ligands were found (cyanidin-3-glucoside, cyanidin-3-rutinoside, cyanidin-3-malonyl-glucoside, and cyanidin-3-dioxalylglucoside), with cyanidin-3-glucoside being the major one. Also, pelargonidin-3-glucoside was found in all the studied cultivars.
Table 3
Anthocyanins identified in the blackberry juices of the BRS Xingu, Guarani, and Xavante cultivars.
Retention time (min) | Molecular and product ions (m/z) | Anthocyanin |
15.274 | 449 − 287 | Cyanidin-3-glucoside |
17.223 | 595 − 287 | Cyanidin-3-rutinoside |
17.924 | 443 − 271 | Pelargonidin-3-glucoside |
22.391 | 535 − 287 | Cyanidin-3-malonyl-glucoside |
23.863 | 593 − 287 | Cyanidin-3-dioxalylglucoside |
When evaluating the effect of the pretreatment with ultrasound on the individual anthocyanins present in the juices (Table 4), it is noted that there was a significant increase (p < 0.05) of the two major anthocyanins (cyanidin-3-glucoside and cyanidin-3-rutinoside) of blackberries in all cultivars studied. However, the different cultivars showed different behaviors regarding the extraction of each anthocyanin.
Table 4
Areas (mV × 106) of individual anthocyanins in the chromatogram for the blackberry juices. Mean value ± standard deviation (n = 3). The samples with * differ from each other using the Student's t-test at a significance level of 5%
Anthocyanin | Cultivar | Treatment |
Cyanidin-3-glucoside | | Ultrasound | Control |
BRS Xingu | 166 ± 4* | 139 ± 3 |
Guarani | 148 ± 5* | 107 ± 11 |
Xavante | 193 ± 13* | 122 ± 3 |
Cyanidin-3-rutinoside | BRS Xingu | 18.9 ± 1.4* | 8.91 ± 1.82 |
Guarani | 14.5 ± 0.3* | 10.4 ± 1.2 |
Xavante | 8.30 ± 0.41* | 5.33 ± 0.03 |
Pelargonidin-3-glucoside | BRS Xingu | 0.432 ± 0.048 | 0.366 ± 0.054 |
Guarani | 0.231 ± 0.001* | 0.219 ± 0.002 |
Xavante | 0.314 ± 0.037 | 0.212 ± 0.024 |
Cyanidin-3-malonyl-glucoside | BRS Xingu | 2.92 ± 0.38 | 2.57 ± 0.14 |
Guarani | 1.88 ± 0.27 | 1.17 ± 0.13 |
Xavante | 2.56 ± 0.19* | 1.56 ± 0.21 |
Cyanidin-3-dioxalylglucoside | BRS Xingu | 6.11 ± 1.07 | 5.72 ± 0.35 |
Guarani | 1.02 ± 0.04 | 0.983 ± 0.133 |
Xavante | 0.132 ± 0.022 | 0.101 ± 0.018 |
As previously reported in the section Effect of ultrasound pretreatment on anthocyanin content of the juice, the cultivar Xavante presented the highest content of extracted anthocyanins, where the area of cyanidin-3-glycoside increased by around 58%, and cyanidin-3-rutinuside had an increase of 55%. For this cultivar, cyanidin-3-dioxalyl-glycoside also increased (65% in relation to the control).
Although the BRS Xingu and Guarani cultivars presented the same increase in total monomeric anthocyanins, the individual anthocyanins of each cultivar increased at different proportions. In the BRS Xingu cultivar, only two anthocyanins differed from the control treatment, increasing 20% for cyanidin-3-glycoside, and 112% for cyanidin-3-rutinoside. As for the Guarani cultivar, the increase was 38% and 40% for cyanidin-3-glycoside and cyanidin-3-rutinoside, respectively. For this same cultivar, there was only a slight increase of 6% in perlagonidine-3-glycoside.
Thus, it can be seen that each cultivar presented a particular behavior regarding the content of anthocyanin increasing in the juice. However, although the anthocyanins of the three cultivars increased in different proportions, the two main anthocyanins for all samples increased significantly for the juices of sonicated blackberries.